HIGHER SOCIAL MOBILITY AND ITS IMPLICATIONS FOR PUBLIC POLICIES

Transcription

1 HIGHER SOCIAL MOBILITY AND ITS IMPLICATIONS FOR PUBLIC POLICIES Certain backgrounds demonstrate that Chile presents a high social mobility and that the proportion of people who have improved their condition would exceed the proportion of those who see it get worse. The fact that in our country the income of the poorest has increased more than that of the richest can also be confirmed. Given the high upward mobility in Chile, the static view of traditional inequality indicators does not allow a precise diagnosis of our society. This has implications for the type of public policies that should be applied to uproot poverty. In the first days of April, L&D received the visit of the prominent Canadian professor, Herbert Grübel, who presented us interesting conclusions on his studies on social mobility in Canada and the United States. One of Grübel s central ideas is that when you compare the income level of families from the lowest quintile for different periods of time, information is lost while overlooking the fact that families belonging to the lower income group in one period or another are not necessarily the same ones. Studying the evolution of the condition of lower income families is as relevant as analyzing what happens with households in this segment of the population: Did households of the first quintile remain in this condition over time? The answer to these questions refers precisely to the concept of social mobility. As in the evidence for Canada, we observe that in Chile there is great mobility, that is, it is highly probable that families composing the lower income 20% change over time, which has clear effects on the type of public policies that need to be implemented. In Chile and around the world, the debate has been focused on the degree and evolution of inequality. The entire discussion has taken place on the base of inequality indicators that may turn out to be tremendously deceptive, since they are a photograph of a specific moment, which does not consider the mobility of individuals. Furthermore, considering that inequality is a relative 1

2 indicator, it does not take into account who is in a specific condition, nor any generalized improvement in the life condition of people. This omission could have direct implications on the design and effectiveness of social programs. It is not the same to design a social policy for a group of individuals who remain stagnated for decades in the first quintile, than for families moving between different income quintiles on their own. MEASUREMENT OF THE MOBILITY LEVEL: FURTHER DATA ARE REQUIRED Social mobility can be upward (improvement of the social condition) or downward (worsening of the social condition). Ideally, social mobility is measured through panel surveys, which are longitudinal surveys that follow a family or individual throughout a particular time, and study their transition regarding their income levels in relation to the rest or to a predetermined income line. Chile has made some efforts to consolidate a panel survey; the best known survey is CASEN Panel Then there is CASEN Panel and the one conducted after the 27F earthquake, known as the Post Earthquake Survey However, all of them have been questioned because of their high attrition rate, which (for example) reached 50% in the case of the CASEN Panel Survey i An elevated attrition rate reflects that a high percentage of families who were interviewed in the first data collection of the survey were not available for the subsequent data collections and, therefore, the collected data are incomplete, which ends up impacting the survey s representativeness, especially if the attrition is associated to some of the conditions being measured. Although in Chile efforts have been made to measure mobility through these surveys, the truth is that, for the aforementioned reasons, we do not rely today on the database that would allow estimating it categorically. In this context, we can only work with them while recognizing that the results will not necessarily be perfect, but certainly they will help us get closer to a preliminary diagnosis concerning the mobility rate of our society. Thus, a first challenge for our country lies in having more and better panel-type information available and searching for indicators that help us complement the information collected through the existing surveys. MOBILITY INDICATORS: NEW PERSPECTIVE ON POVERTY AND INEQUALITY Grübel (2015) ii reports the existence of upward mobility in families belonging to the poorest 20% of the Canadian population in The author observes that many of these families did no longer belong to the first quintile, thereby increasing their incomes. Between 1990 and 2009 only 13% of the first quintile individuals in 1990 was still in the first quintile, 21% passed to the second quintile, 2

3 24% to the third, 21% to the fourth and 21% to the fifth. Likewise, some families who initially belonged to the richest 20% suffered downward mobility. In the same period, 64% of the individuals belonging to the fifth quintile remained in this quintile, 16% passed to the fourth quintile, 7% to the third, 6% to the second and 7% to the first quintile. Additionally, considering the Canadian families who belonged to the poorest 20% and the richest 20% in 1990, the professor calculates the 20/20 ratio and observes a higher reduction of this inequality indicator, if compared to the traditional one which defines quintiles for each year. This means that inequality among Canadian households who were poorer 20 years ago in relation to the richer ones (also 20 years ago) did experience a significant reduction over time, even though the traditional indicator, which does not consider social mobility, does not show it. Results for Canada allow concluding that in a society with high mobility, the traditional inequality calculation does not capture social mobility and it is rather a quite misleading indicator, because it does not only omit income differences that are inherent to the cycle of life, but also economical improvements of those who are no longer poor. For example, in a society with high social, mobility, the traditional 20/20 ratio will compare, in every new measurement, a new group of young people without experience who have just finished their education, those temporarily unemployed and even retired, with adults in the peak of their professional careers, which is naturally an irrelevant comparison from the point of view of the income inequality and for the design of public policies. The surprising and relevant factor is that through the CASEN Panel , the only data available for this type of calculations in our country, we observe the same results of Professor Grübel in Chile: following all families belonging to the different quintiles of 1996, we observe that these groups change over time: those who belonged to the first quintile improved their condition and others who belonged to the richest 20% worsened it. This panel survey shows that between 1996 and 2006, only 42% of the people belonging to the first quintile remained in the first quintile, 27% passed to the second, 18% to the third, 10% to the fourth and 3% to the fifth. As for the richest 20%, in that same period, 56% remained in the fifth quintile, 20% passed to the fourth, 6% to the third, 12% to the second and 5% to the first. In brief, there is enough data to conclude that our country presents high mobility, and that the proportion of individuals improving their condition seems to exceed the proportion of those who see it get worse (Table 1). 3

4 Autonomous Income Quintile in 1996 PUBLIC ISSUES IN 10 YEARS, 58% OF THE PEOPLE WHO BELONGED TO THE FIRST QUINTILE IN 1996 IMPROVED THEIR CONDITION Table 1: People s mobility among different autonomous income quintiles from 1996 to 2006 Autonomous Income Quintile in 2006 I II III IV V TOTAL I 42% 27% 18% 10% 3% 100% II 21% 23% 26% 17% 14% 100% III 12% 21% 23% 27% 17% 100% IV 6% 12% 25% 27% 29% 100% V 5% 12% 6% 20% 56% 100% Source: LyD based on CASEN Panel Survey Chart 1 shows this high mobility in the inequality measurement: our 20/20 inequality index is lower if we follow the families who belonged to the poorest 20% in If we use the composition of the 1996 quintiles in all periods, we obtain for 2006 a 20/20 ratio of 3.5, far below what we would obtain with quintiles of each year (8.7). This result is consistent with those of other studies showing that there is social mobility in Chile; iii for example, the study of Claudio Sapelli (2991), who uses a cohort analysis, concludes that in Chile both the intragenerational (of a same person over time) and the intergenerational social mobility (among different generations) have increased upwardly. iv In short, given the high upward mobility in Chile, the static perspective of the traditional inequality indicators does not allow obtaining an accurate diagnosis of our society. TRADITIONAL INEQUALITY INDICATORS DO NOT CAPTURE MOBILITY IMPROVEMENTS Chart 1: Evolution of the 20/20 ratio with 1996 quintile vs. quintiles of each year 16,0 14,0 12,0 10,0 8,0 6,0 4,0 2,0 0,0 14,0 12,2 8,7 5,9 3, /20 ratio with quintiles of /20 ratio with quintiles of each year 4

5 Source: LyD based on CASEN Panel Survey HIGHER INCOME FOR EVERYONE, ESPECIALLY THE POOREST Thus, Grübel (2015) finds that the income of the poorest in Canada has increased more than that of the richest. The CASEN Survey allows coming to the same conclusions in Chile. For this, we used the total income per capita per household in Chilean pesos of 2013 of the different quintiles, and we concluded that, although the richest 20% has increased their income significantly, the poorest 20% has also increased their incomes. More specifically, a positive evolution of the income for all households is evidenced. Between 1990 and 2013, the annualized income growth rate was higher for the 20% of lower resource households. For this period, the rate was 8.2% for the first quintile, 7.4% for the second quintile, 6.8% for the third, 6.2% for the fourth and 5.8% for the fifth quintile (Chart 2). THE FIRST QUINTILE HAS INCREASED THEIR INCOME AT AN ANNUAL RATE OF 8.2%, 1.5 TIMES MORE THAN THE INCREASE REGISTERED FOR THE FIFTH QUINTILE Chart 2: Annualized growth rate of the total income per capita between 1990 and ,5% 8,0% 7,5% 7,0% 8,2% 7,4% 6,8% 6,5% 6,0% 6,2% 5,5% 5,3% 5,0% i ii iii iv v Source: LyD based on CASEN Surveys. GROWTH OF THE MIDDLE CLASS As for the high rate of social mobility, the World Bank concluded in 2012 that 60.7% of the population in Chile has improved its condition between 1995 and 2010 v, while in the same period only 42.9% of the population of the region had improved. Therefore, the World Bank uses an absolute definition that defines the social condition as follows: poor, when the daily income per 5

6 person is less than US$4; vulnerable, between US$4 and US$10 per day; for the middle class, a threshold of US$10 to US$50; and high income population, over US$50 per day. If we apply the definition used by the World Bank to the CASEN surveys, it is possible to conclude that the evolution of social mobility in our country, from 1996 to date, has gone mainly upward. More specifically, as from the 90 s, the poverty rate in our country has decreased in a significant and sustained manner, from 34.1% to 4.5% (from 4,415,949 to 781,090 people). Likewise, the proportion of vulnerable households has also reduced considerably, from 39% to 29.1% (from 5,054,335 to 5,029,194 people). THE MIDDLE CLASS HAS MORE THAN DOUBLED BETWEEN 1990 AND 2013, FROM 24.2 TO 57.8% OF THE POPULATION Chart 3: Evolution of the proportion of families in Chile, % 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 2,7% 5,6% 6,3% 7,4% 8,6% 24,2% 39,0% 34,1% 37,9% 38,8% 17,8% Source: LyD based on CASEN Survey and World Bank definition. On the contrary, the middle class and high-income population have increased significantly their presence in the society, from 24.2% to 57.8% (from 3,140,784 to 9,977,581 people) and from 2.7% to 8.6% (from 346,435 to 1,485,252 people), respectively (Chart 3). The middle class is, no doubt, the group that has most increased its presence between 1990 and IMPLICATIONS OF A HIGH SOCIAL MOBILITY FOR PUBLIC POLICIES The presence of social mobility has an impact on the design and execution of public policies: if there is a mobile society and social programs are rigid, the social policy does not adjust to the 6 49,1% 51,1% 35,9% 34,5% 57,8% 29,1% 8,7% 7,0% 4,5% Poor Vulnerable Middle Class High Incomes

7 needs of the families, it loses focalization and thus becomes less effective. Another point to consider is dealing with mobility sources and revise policies that ensure these events. A clear example of the above is the score of the Social Protection Card (Ficha de Protección Social - FPS), which has been frozen since 2012 and currently makes the identification of the lower income population more difficult, which is not necessarily the same than a couple of years ago. If a family improves its condition, the frozen score continues to include it among those who require social aid, so that social spending is increased and loses the focus on those who no longer require it. Therefore, one way of keeping spending focalized is to create less resistance to reevaluation and review of the scores; for example, programs should implement a gradual withdrawal of benefits as the family income changes. So, facing a society with high mobility and more middle class implies a different approach in the policy to uproot poverty, since the current design is based on the idea of permanent poverty, where the individuals social condition is maintained. Instead, high mobility implies that being in a poverty condition would rather respond to a transitory reality and, therefore, another type of tools is required, like labor force adjustment programs, where people play the leading roles of their self-improvement. i Bendezú, L.A. Denis and J.R. Zubizarreta (2007). Análisis de la atrición de la muestra en la encuesta panel CASEN. Social Observatory, Universidad Alberto Hurtado. Santiago de Chile, October ii Grübel, H. (2015). Income Mobility or Income Equality? Simon Fraser University, January iii Torche and Wormald (2004). Estratificación y movilidad social en Chile: entre la adscripción y el logro. It shows that mobility is relevant. Nuñex and Risco (2005). Movilidad intergeneracional del ingreso en Chile. It shows that the intergenerational mobility levels have increased. On the other hand, Nuñex and Miranda (2011). Intergenerational income Mobility in a Less-Developed, High Inequality Context: The Case of Chile finds that intergenerational elasticity is high in relation to the international evidence and finds greater mobility in the younger cohorts. iv Sapelli, C. (2011). Chile: Más equitativo? Una mirada distinta a la distribución del ingreso, la movilidad social y la pobreza en Chile. UC Editions, July v Ferreira, Messina, Rigolini, López-Calva, Lugo and Renos. La movilidad económica y el crecimiento de la clase media en América Latina. World Bank. Washington D.C